DynamoDB

Use this skill for DynamoDB-specific data modeling and access-pattern decisions.

Start With Access Patterns

• What queries must be fast and frequent.
• Which entities are read together.
• Expected write distribution and burst patterns.
• Whether global secondary indexes are truly needed.
• Retention, TTL, and item-size constraints.

Design Principles

• Model for access patterns first, not for relational elegance.
• Partition-key quality is central to scale and cost.
• Composite keys should reflect real retrieval needs.
• Secondary indexes are powerful but not free.
• Single-table design should simplify reads, not become a puzzle nobody can maintain.

What To Surface

• Candidate partition and sort key designs.
• Hot-partition or uneven-distribution risks.
• Index tradeoffs and cost implications.
• Where denormalization helps or hurts.
• Questions that must be answered before locking the model.

Common Mistakes

• Leading with entity modeling instead of query modeling.
• Creating GSIs for every imagined future query.
• Ignoring write hotspots.
• Forcing relational patterns into DynamoDB without adaptation.